Scholarlink

Background <p>Laryngotracheal stenosis (LTS) poses life-threatening risks, with diagnostic delays exacerbating morbidity. Non-invasive diagnostic modalities lack standardized validation, particularly in resource-limited settings.</p> Objectives <p>To synthesize diagnostic test accuracy (DTA) evidence for non-invasive LTS modalities (Computed Tomography [CT], CT Virtual Bronchoscopy [CTVB], Spirometry, Magnetic Resonance Imaging [MRI], Ultrasonography, and Conventional Radiography) in adolescents and adults (≥ 12 years), encompassing symptomatic and high-risk asymptomatic populations, and develop evidence-based diagnostic algorithms.</p> Methods <p>Following PRISMA-DTA guidelines, and a PROSPERO-registered protocol (CRD420251044416), 14 studies (30 test evaluations) were identified via systematic searches across eight databases (2000–2025). Bivariate random-effects meta-analysis derived pooled sensitivity/specificity (characterized by HSROC curves), NLR/PLR, DOR, PPV, and NPV. Risk of bias (QUADAS-2), evidence certainty (GRADE), heterogeneity (I²), and pre-specified subgroup/sensitivity analyses were conducted. Laryngotracheobronchoscopy (LTB) or CT served as the reference standard.</p> Results <p>Pooled sensitivity was 91.5% (95% CI: 88.7–93.6%) and specificity 90.8% (79.7–96.1%), demonstrating excellent rule-out utility (NLR: 0.151) and strong discriminatory power (DOR: 53.123; *<i>p</i>*&lt;0.001), despite substantial heterogeneity (I²&gt;90%). Moderate-certainty evidence identified EDI &gt; 50 spirometry (sensitivity 92% [86–96%], specificity 94% [90–96%]) as optimal for screening (minimizing false positives), and CTVB diameter measurements (sensitivity 92% [85–96%], specificity 88% [80–93%]) for confirmation. MRI exhibited critically low specificity (6–46%; very low certainty). Accuracy improved with age (DOR + 10%/decade; *<i>p</i>*&lt;0.001) and in non-tertiary settings (10.7-fold DOR advantage; *<i>p</i>*&lt;0.001). High prevalence (≥ 50%) doubled false negatives.</p> Limitations <p>High heterogeneity, risk of bias (64% studies), spectrum bias (64% surgical cohorts), and absence of LMIC data.</p> Conclusions <p>EDI &gt; 50 spirometry and CTVB measurements form the evidence-based diagnostic foundation; MRI is contraindicated. Implementation reduces diagnostic delays by 23–41% and unnecessary bronchoscopies by 58%. Policy priorities include integrating standardized protocols into guidelines, scaling spirometry access in LMICs, and prospective validation in non-tertiary/LMIC settings.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Diagnostic accuracy of non-invasive modalities for laryngotracheal stenosis: A systematic review and meta-Analysis

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2025年10月24日

查看全文
  • Gerhard Johan Klopper,
  • Oladele Vincent Adeniyi

摘要

Background

Laryngotracheal stenosis (LTS) poses life-threatening risks, with diagnostic delays exacerbating morbidity. Non-invasive diagnostic modalities lack standardized validation, particularly in resource-limited settings.

Objectives

To synthesize diagnostic test accuracy (DTA) evidence for non-invasive LTS modalities (Computed Tomography [CT], CT Virtual Bronchoscopy [CTVB], Spirometry, Magnetic Resonance Imaging [MRI], Ultrasonography, and Conventional Radiography) in adolescents and adults (≥ 12 years), encompassing symptomatic and high-risk asymptomatic populations, and develop evidence-based diagnostic algorithms.

Methods

Following PRISMA-DTA guidelines, and a PROSPERO-registered protocol (CRD420251044416), 14 studies (30 test evaluations) were identified via systematic searches across eight databases (2000–2025). Bivariate random-effects meta-analysis derived pooled sensitivity/specificity (characterized by HSROC curves), NLR/PLR, DOR, PPV, and NPV. Risk of bias (QUADAS-2), evidence certainty (GRADE), heterogeneity (I²), and pre-specified subgroup/sensitivity analyses were conducted. Laryngotracheobronchoscopy (LTB) or CT served as the reference standard.

Results

Pooled sensitivity was 91.5% (95% CI: 88.7–93.6%) and specificity 90.8% (79.7–96.1%), demonstrating excellent rule-out utility (NLR: 0.151) and strong discriminatory power (DOR: 53.123; *p*<0.001), despite substantial heterogeneity (I²>90%). Moderate-certainty evidence identified EDI > 50 spirometry (sensitivity 92% [86–96%], specificity 94% [90–96%]) as optimal for screening (minimizing false positives), and CTVB diameter measurements (sensitivity 92% [85–96%], specificity 88% [80–93%]) for confirmation. MRI exhibited critically low specificity (6–46%; very low certainty). Accuracy improved with age (DOR + 10%/decade; *p*<0.001) and in non-tertiary settings (10.7-fold DOR advantage; *p*<0.001). High prevalence (≥ 50%) doubled false negatives.

Limitations

High heterogeneity, risk of bias (64% studies), spectrum bias (64% surgical cohorts), and absence of LMIC data.

Conclusions

EDI > 50 spirometry and CTVB measurements form the evidence-based diagnostic foundation; MRI is contraindicated. Implementation reduces diagnostic delays by 23–41% and unnecessary bronchoscopies by 58%. Policy priorities include integrating standardized protocols into guidelines, scaling spirometry access in LMICs, and prospective validation in non-tertiary/LMIC settings.